/// <summary>
/// Распознать текст нечитаемого PDF-файла.
/// </summary>
/// <param name="page">Нечитаемая PDF-страница.</param>
/// <returns>Распознанный текст со страницы.</returns>
private static string RecognizePageText(PdfPage page)
{
var options = PdfDrawOptions.Create();
options.BackgroundColor = new PdfRgbColor(255, 255, 255);
options.HorizontalResolution = 200;
options.VerticalResolution = 200;
using var memoryStream = new MemoryStream();
page.Save(memoryStream, options);
using var engine = new TesseractEngine(@"tessdata\fast", "rus+eng", EngineMode.LstmOnly);
using var img = Pix.LoadFromMemory(memoryStream.GetBuffer());
using var recognizedPage = engine.Process(img);
return(recognizedPage.GetText());
}
static void Main(string[] args)
{
// Changes \ to / in the path string
string oldstr = @"\";
string newstr = @"/";
string pdf_path_original = string.Empty;
// Tells the file path
Console.WriteLine("FILE PATH: ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("ファイル パス: ");
pdf_path_original = Console.ReadLine();
string pdf_path = pdf_path_original.Replace(oldstr, newstr);
// Tells the file name
Console.WriteLine(".pdf`s Name: ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("ファイルの名前は: ");
string pdf_file_name = Console.ReadLine();
string path_n_name = pdf_path + "/" + pdf_file_name + ".pdf";
// Tells the file alignment
Console.WriteLine("Text alignment: ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("テキスト配置: ");
Console.WriteLine("1: Left (左), and Right (右), or 2: Middle (真ん中)");
string caseSwitch = Console.ReadLine();
Console.WriteLine();
// Tells the file start page and end page
Console.WriteLine(".pdf`s Initial Page ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("最初のページは: ");
string startpage_string = Console.ReadLine();
int startpage = Int32.Parse(startpage_string);
Console.WriteLine(".pdf`s Last Page ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("最後のページは: ");
string endpage_string = Console.ReadLine();
int endpage = Int32.Parse(endpage_string);
// Creates blank pdf files
for (int i = startpage; i <= endpage; i++)
{
string convi = i.ToString();
// Creates a temp folder for the pdf files
System.IO.Directory.CreateDirectory(pdf_path + "/" + "temp");
System.IO.FileStream pdf_creator = new System.IO.FileStream(pdf_path + "/" + "temp" + "/" + convi + ".pdf", System.IO.FileMode.Create);
pdf_creator.Close();
}
// Tells the attributes from the new pdf files, and the original pdf source
iTextSharp.text.pdf.PdfReader reader = null;
iTextSharp.text.Document sourceDocument = null;
iTextSharp.text.pdf.PdfCopy pdfCopyProvider = null;
iTextSharp.text.pdf.PdfImportedPage importedPage = null;
reader = new iTextSharp.text.pdf.PdfReader(path_n_name);
sourceDocument = new iTextSharp.text.Document(reader.GetPageSizeWithRotation(startpage));
sourceDocument.Open();
// Creates a .docx to receive the pdf's text
Spire.Doc.Document word_doc = new Spire.Doc.Document();
// Word doc formatting
Spire.Doc.Section word_doc_section = word_doc.AddSection();
Spire.Doc.Documents.Paragraph word_doc_paragraph = word_doc_section.AddParagraph();
Spire.Doc.Documents.Paragraph word_doc_paragraph_page = word_doc_section.AddParagraph();
// Update those blank pdf files, inserting the copied pages into it
try
{
for (int i = startpage; i <= endpage; i++)
{
string convi = i.ToString();
pdfCopyProvider = new PdfCopy(sourceDocument, new System.IO.FileStream(pdf_path + "/" + "temp" + "/" + convi + ".pdf", System.IO.FileMode.Append));
sourceDocument.Open();
importedPage = pdfCopyProvider.GetImportedPage(reader, i);
pdfCopyProvider.AddPage(importedPage);
}
sourceDocument.Close();
reader.Close();
}
// ERROR
catch (Exception ex)
{
Console.WriteLine("Error! ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("エラー ! ");
throw ex;
}
// Collects the text without furigana from the listed pdf files
switch (caseSwitch)
{
// case 1 reffers to the left and right alignments of the pdf text
case "1":
Console.WriteLine();
for (int i = startpage; i <= endpage; i++)
{
// the following refers to the int counter of pages being converted into string
string convi = i.ToString();
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("今のページ： " + convi);
Console.WriteLine("Current Page： " + convi);
// the following refers to the bitmiracle api pdf to get the texts
using (BitMiracle.Docotic.Pdf.PdfDocument pdf_1 = new BitMiracle.Docotic.Pdf.PdfDocument(pdf_path + "/" + "temp" + "/" + convi + ".pdf"))
{
BitMiracle.Docotic.Pdf.PdfPage page = pdf_1.Pages[0];
foreach (PdfTextData data in page.GetWords())
{
if (data.FontSize > 6 && data.Position.X < 600)
{
string text = data.Text;
text.TrimEnd();
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine(text);
//word_builder.Writeln(text);
word_doc_paragraph.AppendText(text);
}
}
foreach (PdfTextData data in page.GetWords())
{
if (data.FontSize > 6 && data.Position.X > 600)
{
string text = data.Text;
text.TrimEnd();
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine(text);
word_doc_paragraph.AppendText(text);
}
}
}
// the following lines reffers to the space between pages of the pdf text
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
// the followin reffers to the extra lines on word text
word_doc_paragraph.AppendText(" ");
word_doc_paragraph.AppendText("CURRENT PAGE： " + convi);
word_doc_paragraph = word_doc_section.AddParagraph();
word_doc.Sections[0].Paragraphs[i].AppendBreak(BreakType.PageBreak);
}
break;
// case 2 reffers to the alignment of the pdf text that is centralized
case "2":
Console.WriteLine();
for (int i = startpage; i <= endpage; i++)
{
// the following refers to the int counter of pages being converted into string
string convi = i.ToString();
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("今のページ： " + convi);
Console.WriteLine("Current Page： " + convi);
// the following refers to the bitmiracle api pdf to get the texts
using (BitMiracle.Docotic.Pdf.PdfDocument pdf_1 = new BitMiracle.Docotic.Pdf.PdfDocument(pdf_path + "/" + "temp" + "/" + convi + ".pdf"))
{
BitMiracle.Docotic.Pdf.PdfPage page = pdf_1.Pages[0];
foreach (PdfTextData data in page.GetWords())
{
if (data.FontSize > 6)
{
string text = data.Text;
text.TrimEnd();
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine(text);
word_doc_paragraph.AppendText(text);
}
}
}
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
Console.WriteLine();
word_doc_paragraph.AppendText(" ");
word_doc_paragraph.AppendText("CURRENT PAGE： " + convi);
word_doc_paragraph = word_doc_section.AddParagraph();
word_doc.Sections[0].Paragraphs[i].AppendBreak(BreakType.PageBreak);
}
break;
default:
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("error! (エラー)");
Console.ReadKey();
break;
}
// The following refers to creating a .docx file, opening up the file and deleting the temp folder
word_doc.SaveToFile(pdf_path + "/" + pdf_file_name + ".docx", FileFormat.Docx);
System.IO.Directory.Delete(pdf_path + "/" + "temp", true);
try
{
System.Diagnostics.Process.Start(pdf_path + "/" + pdf_file_name + ".docx");
}
catch
{
Console.WriteLine("Error! ");
Console.OutputEncoding = Encoding.GetEncoding(932);
Console.WriteLine("エラー ! ");
}
}